We have developed a new technology called Gene Identification Signature Pair-end-diTag (GIS-PET), which captures precisely and joins the 5' and 3' most 18 base pair of any clone insert in a library. The resultant "ditags" are signatures of the boundaries of any cDNA or genomic fragment within the library. When coupled with advanced sequencing technologies, GIS-PET can annotate an entire transcriptome 300-500 fold faster and less expensively than contemporary cloning and sequencing. We have captured the transcriptome snapshot of the MCF-7 breast cancer cell line to >500,000 full length cDNA equivalents and have captured the state of the transcriptome of this breast cancer cell line with unprecedented resolution. With this approach, we found that we can identify, again with precision, candidate fusion transcripts where the beginning of one gene is fused with the end of another. These fusion transcripts are cancer-specific and can function as biomarkers, as monitors of treatment effect, and as targets for new therapeutics. We therefore propose to expand and validate this technology on primary breast cancers and to confirm the significant list of candidate fusion transcripts already identified in the MCF-7 cell line. We will clone and sequence all the fusion transcripts accessible to these technologies and will assess if their presence in primary breast cancers is associated with cancer behavior and clinical outcome. We also propose to optimize the GIS-PET technology to genomic DNA so as to map the precise location of genomic rearrangements in cancer cells. When put together, the combined power of the technology can uncover the complete rearrangement map of any cancer genome. Because we have brought this technology beyond conceptualization, we are submitting this application as an R33. [unreadable] [unreadable] Relevance to public health: Cancer is a genetic disease and the understanding of the genetic abnormalities in cancer has been shown to uncover new biomarkers and new treatments in the management of the disease. We have developed a new technology that can map the consequences of all the important abnormal rearrangements in a cancer cell. We propose to perfect this technology and to apply it to human breast cancer tissues. [unreadable] [unreadable] [unreadable]